Download Conservation of surface epitopes in Pseudomonas aeruginosa outer

261
FEMS Microbiology Letters 113 (1993) 261-266
© 1993 Federation of European Microbiological Societies 0378-1097/93/$06.00
Published by Elsevier
FEMSLE 05664
Conservation of surface epitopes in Pseudomonas
aeruginosa outer membrane porin protein OprF
Nancy L. Martin 4, Eileen G. Rawling 4, Rebecca S.Y. Wong
and R o b e r t E.W. H a n c o c k .,a
a Mac Rosok b
a Department of Microbiology, University of British Columbia, Vancouver, B.C., Canada V6T 1Z3 and
b Bristol Myers Squibb Pharmaceutical Research Division, 3005 First Avenue, Seattle, Washington, USA
(Received 28 June 1993; revision received 11 August 1993; accepted 17 August 1993)
Abstract: The outer membrane proteins of several prominent bacterial pathogens demonstrate substantial variation in their surface
antigenic epitopes. To determine if this was also true for Pseudomonas aeruginosa outer membrane protein OprF, gene sequencing
of a serotype 5 isolate was performed to permit comparison with the published serotype 12 oprF gene sequence. Only 16 nucleotide
substitutions in the 1053 nucleotide coding region were observed; none of these changed the amino acid sequence. A panel of 10
monoclonal antibodies (mAbs) reacted with each of 46 P. aeruginosa strains representing all 17 serotype strains, 12 clinical isolates,
15 environmental isolates and 2 laboratory isolates. Between two and eight of these mAbs also reacted with proteins from
representatives of the rRNA homology group I of the Pseudomonadaceae. Nine of the ten mAbs recognized surface antigenic
epitopes as determined by indirect immunofluorescence techniques and their ability to opsonize P. aeruginosa for phagocytosis.
These epitopes were partially masked by lipopolysaccharide side chains as revealed using a side chain-deficient mutant. It is
concluded that OprF is a highly conserved protein with several conserved surface antigenic epitopes.
Key words: Porin; pseudomonas aeruginosa; Outer membrane; OprF; Epitope
Introduction
One of the strategies utilized by successful
pathogens to evade the immune response is variation of surface antigen structure. For outer membrane proteins this can involve either mutations
[1], or more complex genetic events such as rearrangements promoted by frameshifts [2]. The ma-
* Corresponding author. Tel.(604) 822-2682, Fax (604) 8224040.
jor outer membrane proteins of Gram negative
bacteria have been shown, by either monoclonal
antibody (mAb) reactivity or DNA sequence analysis to vary within individual strains of many
species. Thus the outer membrane porin proteins
from Escherichia coli [3], Haemophilus influenzae
[4], Chlamydia trachomatis [5] and Neisseria sp. [6]
have all been shown to have variations in surface
antigenic structure, whereas internal epitopes are
generally more highly conserved.
OprF is the major porin protein in P. aeruginosa that forms channels through the outer mem-
262
brane that are large enough to accommodate diand tri-saccharides [9]. In addition, a second
function in both outer membrane and cell structure and stability was discovered, in that OprFdeficient mutants have rounded morphology and
grow well only in high osmolarity media [10,11].
Studies utilizing mAbs specific for OprF have
demonstrated that this protein is potentially useful as a target for diagnostic and immunotherapeutic intervention [12,13]. Indeed, Gilleland and
colleagues [14] have presented data to demonstrate that immunization with O p r F protects
against subsequent Pseudomonas aeruginosa infections.
Previous studies with two mAbs directed
against Pseudomonas aeruginosa protein OprF
demonstrated the conservation of two surface
epitopes that were present in all P. aeruginosa
strains tested [7]. Furthermore, the oprF gene
cross-hybridized in all strains and demonstrated
rather conserved restriction patterns [8]. However, the restricted number of mAbs tested, and
the observation of variation in surface antigenic
structure in other bacterial pathogens, led us to
examine here whether OprF was an unusual example of an outer membrane protein with general conservation of surface epitopes.
Materials and Methods
Strain and media
P. aeruginosa PAO strains utilized included
wild type strain H103 (IATS serotype 5; Hancock
and Carey, 1979), its OprF::£1 mutant H636 [10]
and H692, a mutant unable to produce A and B
band LPS side chains (a gift from Dr. J. Lam,
University of Guelph, Canada). Strain M2 [13]
was an IATS serotype 5 strain (i.e., the same
serotype as strain H103). Other Pseudomonadaceae strains used were described previously
[7,16,17]. Growth media included Mueller Hinton
Broth (all strains) and proteose peptone No. 2
(PP2) without (H103, M2) or with (H636) the
addition of 200 mM NaC1 and 500 /xg/ml streptomycin. Strain H692 was grown on 1% Bactotryptone, 0.5% yeast extract, 0.5% NaC1.
Immunological techniques
Enzyme-linked immunoabsorbent assays were
performed with wells coated with either whole
cells, isolated outer membranes or purified OprF,
following procedures previously described by
Mutharia and Hancock [17]. Cell surface indirect
immunofluorescence labelling was carried out on
mid-logarithmic or stationary phase cells collected by centrifugation from 1.5 ml of culture.
The cells were washed twice with PBS by centrifugation and resuspended in 200 /.tl of PBS.
This suspension was smeared on a glass slide
(that in some experiments was precoated with 1
m g / m l poly-L-lysine) and allowed to air dry, then
fixed in 100% ethanol. The slides were then
processed following the method of Hofstra et al.
[18], with minor modifications as described by
Mutharia and Hancock [17]. The slides were examined with a Zeiss microscope fitted with a
condenser for fluorescence microscopy and containing a halogen lamp and suitable filters for
emission of fluorescein isothiocyanate at 525 nm.
Opsonized phagocytosis of P. aeruginosa was
performed using a modification of the procedures
of Battershill et al. [19]. Mouse peritoneal
macrophages from 2 - 4 months old B6D2(fl) mice
were incubated in 8-well chamber slides (NUNC,
Naperville, IL) overnight and rinsed twice prior
to addition of bacteria, thereby eliminating the
cytospin step. Antibody concentrations used were
standardized by ELISA with purified OprF. Experiments were carried out on 3 different days,
the results from each day were tabulated and
subjected separately to the Mann-Whitney test to
determine a significant difference from the control (cells opsonized with mAb MA1-3).
Colony immunoblotting [7] and Western immunoblotting [17] were carried out as described
previously.
DNA sequencing
The oprF gene of P. aeruginosa strain H103
[11] was sequenced as described previously using
oligonucleotide primers [8], except that the protocols of Applied Biosystems Inc. (ABI) were followed for sequencing with an ABI Model 373
automated fluorescence DNA sequenator. The
263
the oprF gene from a serotype 12 isolate that had
been sequenced (Genbank accession number
P13794) by Duchene et al. [20], whereas the majority of strains, including our laboratory wild
type strain H103 and strain M2, lacked this Kpnl
site. To determine the extent of variation of oprF
sequences in P. aeruginosa, the cloned gene [11]
from strain H103 was sequenced in this study
(Genbank accession number M94078). Only 16
nucleotide pair changes were observed relative to
the serotype 12 sequence. Nine of these were
C --) T (at nucleotide positions 69, 175, 225, 486,
624, 650, 789, 954 and 1,029 of the open reading
frame), four were T---> C (at positions 228, 267,
516, and 837), two were T --) G (at positions 150
and 204), and one was G--) A (at position 930).
All of these changes were silent and thus the two
amino acid sequences were identical.
MA7-4,5,7
MA7-2
MA7-3
MA7-1
MA4-4,MA7-8
M
I
M
I
M
l
I__
50
7-----150
100
F i g . 1. L i n e a r m a p o f O p r F
approximate
topes based
positions
on data
from
MAS-8
M
I
[
I
200
I
250
I
300
with positions of the methionine
(M) and cysteine (C) residues
the
CCCC
~lf
MA7-6
indicated.
of the
N. Martin,
1 9 9 2 a n d [15l. T h e n u m b e r s
Above
monoclonal
the map are
antibody
Ph.D.
epi-
Thesis U.B.C.,
refer to amino acid positions.
D N A sequence is deposited in Genbank under
the accession number M94078.
Results
Antigenic conservation of OprF in Pseudomonas
aeruginosa
Sequence conservation of OprF in Pseudomonas
aeruginosa
A panel [15] of ten mAbs recognizing diverse
epitopes of OprF (Fig. 1) were reacted with a
wide variety of P. aeruginosa strains using the
colony immunoblot procedure [7]. All isolates of
P. aeruginosa, except for the negative control
strain H636 oprF::O, reacted on colony immunoblots with all 10 mAbs (Table 1), as confirmed in selected cases by Western immunoblot-
Previous experiments [8] probing, by Southern
hybridization, the oprF gene of 59 P. aeruginosa
clinical and serotyping strains indicated only one
restriction endonuclease polymorphism observed
in 2 strains which demonstrated an additional
K.pnl site in the carboxy terminal half of the oprF
gene. This unusual pattern was also observed for
Table
1
Monoclonal
antibody reactivity of various Pscudomonads
Strains a
a The designation
for
homology group I
Reactivity b
P. aeruginosa ( 4 6 i s o l a t e s )
P. syringae A T C C 1 9 3 1 0 x
P. fluoresccens A T C C 135253P.fluorescens O E 2 8 . 3
P. putida A T C C 126333P. stutzerii A T C C 1 7 5 7 8 T
P. aureofaciens A T C C 139853P. chloraphis A T C C 9 4 4 6 T
A. vinelandii O P
b + + Reaction
of rRNA
superscript
on Western
P. aeruginosa
OprF;
MA4-4
MAS-8
MA7-1
MA7-2
MA7-3
MA7-4
MA7-5
MA7-6
MA7-7
+ +
+ +
+ +
+ +
+ +
+ +
+ +
+ +
+ +
+ +
+ +
-
-
+ +
-
-
-
+ +
-
+ +
+
-
-
-
+ +
-
_
_
+ +
_
+
_
+ +
+ +
_
+ +
-
-
+ +
+
+
+
+
+ +
+ +
+
-
-
+
+ +
-
-
+ +
+ +
-
-
-
+
+ +
-
+ +
-
+
+ +
-
-
-
+
+ +
-
+ +
-
+
+ +
-
-
-
-
+
+
-
-
-
+
-
.
_ c
.
.
T i n d i c a t e s t h a t t h i s is t h e t y p e s t r a i n o f t h e s p e c i e s .
immunoblot
-
.
MA7-8
equivalent to OprF of
no reaction. All samples
P. aeruginosa
were heated
strain H692;
+ reaction of lesser intensity than that
a t 1 0 0 ° C f o r 15 m i n in s o l u b i l i z a t i o n
2-mercaptoethanol,
prior to SDS-PAGE.
R e s u l t s f o r t h e 46 P. aeruginosa s t r a i n s w e r e o b t a i n e d
v a l i d a t e d in s e l e c t e d i n s t a n c e s b y W e s t e r n i m m u n o b l o t .
c Reactivity was observed with a band of lower apparent molecular mass.
reduction
mix without
by colony immunoblot
[7l a n d
264
ting and ELISA studies. The P. aeruginosa strains
tested included the 17 IATS serotype isolates
[17], 15 clinical strains from different countries,
15 environmental isolates, and 3 laboratory isolates. These data thus indicated strong conservation of OprF epitopes in P. aeruginosa.
Table 2
Indirect immunofluorescence
labelling of intact
P. aeruginosa
by O p r F - s p e c i f i c m o n o c l o n a l a n t i b o d i e s a n d t h e c o n t r o l a n t i body MAI-3
Monoclonal antibody
lmmunofluorescence
M2
H103
"
H692
MA4-4
+ +
+ +
+ +
Reactivity with other Pseudomonadaceae
Southern blotting studies with the oprF gene
from P. aeruginosa previously indicated that the
oprF gene was conserved in all Pseudomonadaceae of rRNA homology group I [8]. Two
MA5-8
+ +
-
+ +
MA7-1
+ +
-
+
MA7-5
additional genes have now been sequenced, the
P. syringae pv. syringae ATCC 19310 OprF porin
[8] and the P. fluorescens OE283 plant root adhesion protein [21]. Interestingly, although both have
extensive amino acid homology with P. aeruginosa OprF, the latter lacks the entire cysteinecontaining region of OprF, whereas the former
contains this region [21]. Consistent with this, P.
syringae OprF reacted with mAbs MA4-4 and
MA7-8, which recognize the cysteine disulphide
region [15], whereas P. fluorescens OE283 did not
(Table 1).
Of the Pseudomonadaceae strains examined by
Western immunoblot analysis (Table 1), P. putida
showed the broadest cross-reactivity of any of the
strains examined here, reacting with 8 of the 10
mAbs. However, all strains examined reacted with
2 or more mAbs. The weakest cross-reactivity was
observed with ATCC 13525, the type strain of P.
fluorescens, and with Azotobacter vinelandii.
Three mAbs MA7-2, MA7-6 and MA7-7 represented the most conserved epitopes reacting with
7, 7 and 8 strains, respectively. Conversely, the
epitopes for MA5-8, MA7-5 and MA7-1 were
poorly conserved demonstrating weak reactivity
with 0 to 2 other Pseudomonads.
Surface localization studies. It has been previously demonstrated that surface localized regions
of outer membrane proteins demonstrate the
greatest variability [3,4,22]. Thus, it was possible
that the antigenic conservation described above
reflected conservation of epitopes that were not
exposed on the surface. Therefore, indirect immunfluorescence studies were performed to determine whether or not the OprF epitopes being
studied were surface localized (Table 2). Utilizing
MA7-6
MA7-7
MA7-8
MA1-3 b
H636
-
MA7-2
-
-
+
MA7-3
+ +
+
+
-
MA7-4
+ +
+
+ +
-
+
+
+ +
-
+
-
+
+ +
+
+ +
-
+ +
-
+ +
NT c
NT
a _
-
No fluorescence or only weak fluorescence observed,
+
cells u n i f o r m l y fluorescent, + + cells highly fluorescent.
b Negative control antibody specific for a non surface-localized epitope on outer membrane
protein OprI.
" NT, not tested.
strain M2, we observed labelling of the surfaces
of cells with all antibodies studied except MA7-2
and the negative control [19] antibody MA1-3.
Labelling of strain H103 was consistently weaker
and several antibodies failed to label strain H103.
This appeared to be due in part to masking of
OprF by LPS side chains since strain H692, an
LPS-altered, rough derivative of strain H103 was
better labelled with all mAbs. Strain H636, an
oprF::~2 mutant of strain H103, was not labelled
by any of the mAbs.
As an alternative method of studying surface
binding, each of the mAbs was utilized as an
opsonin for phagocytosis of strain M2 by nonelicited mouse peritoneal macrophages. We confirmed previous data [19] which suggested that
MA5-8 and MA4-4 were opsonic and that MA5-8
was superior to MA4-4 in opsonizing for phagocytosis (Fig. 2). At an antibody dilution of 10 2
MA7-3, MA7-5, MA7-6, MA7-7 and MA7-8 also
consistently opsonized P. aeruginosa for phagocytosis. MAT-2 was no better than the control antibody MA1-3 whereas MA7-1 and MA7-4 resulted
in a significant increase in phagocytosis in 1 out
of 3 experiments. When less diluted (10 -1) antibody was also utilized for phagocytosis, MA7-1
consistently gave significant phagocytosis, while
265
¢
a
3
....
*
*
:
~-
:
ili!
2
*
(0
.
*
:
~
*
ro
0
7-1
7-2
7-3
7-4
7-5
7-6
7-7
7-8
4-4
5-8
Antibody
Fig. 2. E n h a n c e m e n t of phagocytosis of P. aeruginosa strain
M2 by OprF-specific monoclonal antibodies. Results represent the average n u m b e r of bacteria associated per mouse
peritoneal macrophage. Each different shaded bar represents
data from an independent experiment with the control number of bacteria per macrophage (obtained using the negative
control antibody MA1-3) subtracted. The star indicates samples that were significantly different from the control (by
M a n n - W h i t n e y test P < 0.05).
other antibodies yielded the same types of results
as shown in Fig. 2. Overall the data suggested
that the epitope for MA7-2 is not exposed (except
perhaps in the LPS-altered strain H692). However, all other epitopes were apparently surface
exposed.
Discussion
In contrast to several outer membrane proteins found in other bacterial species [2-6], OprF
was demonstrated here to contain a number of
discrete, highly conserved surface antigenic epitopes. The mAbs utilized in this study recognized
epitopes that were largely clustered in the carboxy terminal half of the protein. Nevertheless, a
variety of data including their variable reactivity
with OprF equivalents from other Pseudomonads
(Table 1) and their differential reactivity with
peptides [15; N. Martin, Ph.D. Thesis, U.B.C.,
1992] indicate that they recognized separate epitopes. The reason for conservation of OprF epitopes was not definitively determined in this study.
In other bacteria, it has been suggested that
masking of outer membrane protein surface epitopes by LPS limits their reactivity with antibod-
ies specific for these epitopes [23]. This also appeared evident in this study comparing strain
H103 with its LPS-altered, rough derivative H692
(Table 2). However, this cannot be the only important factor since immunofluorescence labelling of strain H103 was observed with some
mAbs and one of those mAbs MA4-4 [13] protects mice against Pseudomonas infections. Furthermore, cystic fibrosis patient isolates, that
demonstrate rough LPS, were fully reactive with
all 10 mAbs, indicating that these isolates had not
undergone extensive oprF gene mutations.
Pseudomonas species from rRNA homology
group I demonstrated immunologic cross-reactivity with subsets of the 10 OprF-specific mAbs,
indicating that OprF has been conserved across
species barriers. Three of the Pseudomonas
species tested have now had their oprF genes
sequenced. Alignment of the sequences of the
OprF proteins from P. aeruginosa [20], P. syringae [8] and P. fluorescens [21] revealed that the
carboxy-terminal 121 amino acids are highly conserved. Compared to P. aeruginosa, P. syringae
OprF has 101 identical amino acids, 12 conservative substitutions and 8 differences while P. fluorescens OprF [21] has 1 single amino acid deletion, 13 different residues, 19 conservative substitutions and 87 identical amino acids (providing
the cysteine disulphide region is excluded from
this analysis). Only two (MA7-2, MA7-6) of the 7
mAbs specific for this region of OprF, crossreacted strongly with all 3 proteins, mAb MA7-2
did not apparently recognize an epitope that is
well exposed on the surface of P. aeruginosa (Fig.
1). Thus, it seems reasonable to conclude that in
these different species, the surface-localized regions, recognized by 5 of the other 6 mAbs with
epitopes in the carboxy terminal 121 amino acids,
have undergone the greatest antigenic variation.
Consistent with this idea, the surface epitopes of
porin I from Neisseria [24] and FepA from Enterobacteriaceae [22] represent the most variable
portions of these proteins.
Acknowledgements
Financial support for this project came from
the Medical Research Council of Canada. E.R.
266
and R.W. were supported by studentships from
the Canadian Cystic Fibrosis Foundation and
Natural Sciences and Engineering Research
Council of Canada, respectively.
References
1 Jeanteur, D., Lakey, J.H. and Pattus, F. (1991) The bacterial porin superfamily: Sequence alignment and structure
prediction. Mol. Microbiol. 5, 2153-2164.
2 Murphy, G.L., Connell, T.D., Barritt, D.S., Doomey, M.
and Cannon, J.G. (1989) Phase variation of gonococcal
protein II: regulation of gene expression by slipped strand
mispairing of a repetitive DNA sequence. Cell 56, 539-547.
3 Pages, J.M., Pages, C., Bernadac, A. and Prince, P. (1988)
Immunological evidence for differences in the exposed
regions of OmpF porins from Escherichia coli B and K-12.
Molec. Immunol. 25, 555-563.
4 Van Alphen, L., Eijk, P., Van Den Brock, L.G. and
Dankert, J. (1991) Immunochemical characterization of
vaiable epitopes of outer membrane protein P2 of nontypeable Haemophilus influenzae. Infect. Immun. 59, 247252.
5 Zhong, G. and Brunham, R.C. (1990) Immunoaccessible
peptide sequences of the major outer membrane protein
from Chlamydia trachomatis serovar C. Infect. Immun. 58,
3438-3441.
6 Virji, M., Fletcher, J.N., Zak, K. and Heckels, J.E. (1987)
The potential protective effect of monoclonal antibodies
to gonococcal outer membrane protein IA. J. Geu. Microbiol. 133, 2639-2646.
7 Mutharia, L.M. and Hancock, R.E.W. (1985) Characterization of two surface-localized antigenic sites of porin
protein F of Pseudomona aeruginosa. Can. J. Microbiol.
31,381-386.
8 Ullstrom, C,A,, Siehne[, R., Woodruff, W., Steinbach, S.
and Hancock, R.E.W. (1990) Conservation of the gene for
outer membrane protein OprF in the Pseudomonaceae:
Sequence of the Pseudomonas syringae oprF gene. J, Bacteriol. 31,768-775.
9 Bellido, F., Martin, N i . and Siehnel, R.J. (1992) Reevaluation in intact cells of the exclusion limit and role of porin
OprF in Pseudomonas aeruginosa outer membrane permeability. J. Bacteriol. 174, 5196-5203.
10 Gotoh, N., Hirokazu, W., Yoshihara, E., Nakae, T. and
Nishino, T. (1989) Role of protein F in maintaining structural integrity of Pseudomonas aeruginosa outer membrane. J. Bacteriol. 171,983-990.
11 Woodruff, W.A. and Hancock, R.E.W. (1986) Expression
in Escherichia coli and function of Pseudomonas aeruginosa outer membrane porin protein F. J. Bacteriol. 171,
3304-3309.
12 Counts, G.W., Schwartz, R.W., Ulness, B.K., Hamilton,
D.J., Rosok, M.J., Cunningham, M.D., Tam, M.R. and
Darveau, R.P. (1977) Evaluation of an immunoassay test
for rapid identification of Pseudomonas aeruginosa in blood
cultures. J. Clin. Microbiol. 26, 1161-1165.
13 Hancock, R.E.W., Mutharia, L.M. and Mount, E.C.A.
(1985) The immunotherapeutic potential of monoclonal
antibodies against P. aeruginosa protein F. Eur. J. Clin.
Microbiol. 4, 224 227.
14 Gilleland, Jr., H.E., Gilleland, L.B., Hughes, E.E. and
Matthews-Greer, J.M. (1992) Recombinant outer membrane protein F of Pseudomonas aeruginosa elicits antibodies that mediate opsonophagocytic killing, but not
complement-mediated bacteriolysis, of various strains of
P. aeruginosa. Curr. Microbiol. 24, 1-7.
15 Finnen, R.L., Martin, N.L., Siehnel, R.J. Woodruff, W.A..
Rosok, M. and Hancock, R.E.W. (1992) Analysis of structural features of the major outer membrane protein OprF
from Pseudomonas aeruginosa using derivatives created by
TnphoA mutagenesis and subcloning. J. Bacteriol. 174,
4977-4985.
16 Hancock, R.E.W. and Chan, L. (1988) Outer membranes
of environmental isolates of Pseudomonas aeruginosa. J.
Clin. Microbiol. 26, 2423-2424.
17 Mutharia, L.M. and Hancock, R.E.W. (1983) Surface localization of Pseudomonas aeruginosa outer membrane
porin protein using monoclonal antibodies. Infect. Immun.
42, 1027 1033.
18 Hofstra, H., Van Tol, M.J.D. and Dankert, J. (1979)
Immunofluorescent detection of the major outer membrane protein II* in Escherichia coli 026K60. FEMS
Microbiol. Eett. 6, 147-150.
19 Battershill, J , Speert, D.P. and Hancock, R.E.W. (1987)
Use of monoclonal antibodies against protein F of Pseudomonas aeruginosa as opsonins for phagocytosis by
macrophages. Infect. lmmun. 55, 2531-2533.
20 Duchene, M., Schweizer, A., Lottspeich, F., Krauss, G.,
Marget, M., Vogel, K., Von Specht, B. and Domdey, It.
(1988) Sequence and transcriptional start site of the P~eudomonas aeruginosa outer membrane porin protein F gene.
J. Bacteriol. 170, 155-162.
21 De Mot, R., Proost, P., Van Damme, J. and Van Der
Leyden, J. (1992) Homology of the root adhesin of P~'eudomonas fluomscens OE 28.3 with porin F of Pseudomonas aeruginosa and Pseudomonas syringae. Mol. Gen.
Genet. 231,489-493.
22 Rutz, J.M., Abdullah, T., Singh, S.P., Kalve, V.I. and
Klebba, P.E. (1991) Evolution of the ferric enterobactin
receptor in Gram-negative bacteria. J. Bacteriol. 173.
5964-5974.
23 Bentley, A.T. and Klebba, P.E. (1988) Effect of
lipopolysaccharide structure on reactivity of anti-porin
monoclonal antibodies with the bacterial cell surface. J.
Bacteriol. 170, 1063-1068.
24 Sandstron, E.G., Knapp, J.S. and Buchanan, T.B. (1982)
Serology of Neisseria gonorrhoeae: W-antigen serogrouping by coagglutination and protein i serotyping by enzyme
linked immunosorbent assay both detect protein I antigens. Infect. Immun. 35,229-239.